lywcn / dui_vio

Depth uncertainty incorporated VIO

DUI_VIO

A RGB-D camera based Visual-Inertial State Estimator

Videos: run with some data sequences in VCU-RVI

Related Paper

He Zhang and Cang Ye, "DUI-VIO: Depth uncertainty incorporated visual inertial odometry based on an RGB-D camera", IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020. pdf

1. Prerequisites

1.1 Ubuntu and ROS Ubuntu 16.04. ROS Kinetic. ROS Installation additional ROS pacakge

    sudo apt-get install ros-YOUR_DISTRO-cv-bridge ros-YOUR_DISTRO-tf ros-YOUR_DISTRO-message-filters ros-YOUR_DISTRO-image-transport

1.2. Ceres Solver Follow Ceres Installation, remember to make install. (Our testing environment: Ubuntu 16.04, ROS Kinetic, OpenCV 3.3.1)

2. Build DUI_VIO on ROS

Clone the repository and catkin_make:

    cd ~/catkin_ws/src
    git clone https://github.com/rising-turtle/DUI_VIO.git
    cd ../
    catkin_make
    source ~/catkin_ws/devel/setup.bash

3. Build with OpenGV (Optional)

OpenGV is not required in default compilation. It is used to improve the accuracy of 2D-2D feature matches for rotation estimation in the HybridPnP algorithm. To compile with OpenGV, set WITH_OPENGV as ON in CMakeLists.txt and set up "opengv_DIR" accordingly

option(WITH_OPENGV "use opengv" ON) #use opengv or not in hybridPnP 
set(opengv_DIR "/PATH_TO_OPENGV/opengv/build")

4. Demos

Download the bag_files which were collected by a structure-core camera handheld or robot carried. Open a terminal, navigate to the 'DUI_VIO/launch' folder, and launch the dui_vio_sc_run. Open another terminal and play the bag file.

    roslaunch dui_vio_sc_run.launch
    rosbag play YOUR_PATH_TO_DATASET/lab_*.bag 

5. Licence

The source code is released under GPLv3 license.